Scientia Horticulturae 293 (2022) 110726 Contents lists available at ScienceDirect Scientia Horticulturae journal homepage: www.elsevier.com/locate/scihorti Comparative QTL analysis in peach ‘Earlygold’ F2 and backcross progenies Naveen Kalluri a, Iban Eduardo a, b, Pere Arús a, b, * a Centre for Research in Agricultural Genomics (CRAG) CSIC-IRTA-UAB-UB, Campus UAB, Edifici CRAG, Cerdanyola del Vallès (Bellaterra), 08193 Barcelona, Spain b IRTA, Campus UAB, Edifici CRAG, Cerdanyola del Vallès (Bellaterra), 08193 Barcelona, Spain A R T I C L E I N F O A B S T R A C T Keywords: Based on detailed maps, DNA sequences and phenotypic data, there is a great deal of information on the genetics Prunus persica and genomics of ‘Earlygold’, a historical peach cultivar from the US. The F2 between ‘Texas’ almond and ‘Ear­ F2 lygold’ peach (T × E) was used to construct the first saturated peach linkage map that later became the reference Backcross map for the Prunus genus. This population and the first backcross (’Texas’ × ‘Earlygold’) × ‘Earlygold’ (T1E) QTL analysis Comparative mapping yielded information on QTLs for a large number of agronomic traits, and T1E is being used as the basis for Leaf senescence color constructing a set of introgression lines of ‘Texas’ fragments into the ‘Earlygold’ background, currently in progress. This paper describes the construction of a high-density SNP map for ‘Earlygold’ using an F2 population, and the QTL analysis of 24 traits. Results of maps and QTLs are compared with those from the ‘Earlygold’ parent of the T1E map, using the same set of markers and characters. Results show major differences between the two progenies in terms of numbers of markers mapped and the capability of detecting QTLs, with a large increase in the resolution of maps and QTLs when using the F2 progeny compared to the T1E pseudo-testcross. In addition, we provide data on leaf senescence color, studied for the first time in peach, with two consistent QTLs located in the same position as other color-related genes and QTLs. Introduction population was useful for map construction, but only about a half of its progeny was fertile and fruited (Donoso et al., 2016), so a BC1 progeny Peach [Prunus persica (L.) Batsch], is an economically important with ‘Earlygold’ as recurrent parent was obtained, where most in­ stone fruit crop and one of the model species of the Rosaceae family dividuals produced fruit. In this cross, the ‘Texas’ cytoplasm resulted in (Shulaev et al., 2008). Like most cultivated members of the Prunus male sterility in the peach nuclear genetic background unless at least genus, such as almond, apricot, plum and cherry, peach is diploid one of two independent restorer factors from ‘Texas’ were present (2n=2x=16) with a compact and sequenced genome of ~x223C 250 (Donoso et al., 2015). The BC1 progeny and further backcross and Mbp (Verde et al., 2013). It has a self-compatible mating system and a selfing generations have been used as a proof of concept for short intergeneration period of 3–4 years, in contrast to other rosaceous marker-assisted introgression (MAI), a fast strategy to obtain individuals fruit tree species that are usually self-incompatible and require a longer with a single almond introgressed fragment in the BC2 (Serra et al., time for fruiting. Some of the major targets in the current peach breeding 2016). These materials are currently being used to develop an intro­ programs are difficult to meet, such as extended shelf life, better fruit gression line (IL) collection of almond fragments in the ‘Earlygold’ quality and enhanced disease resistance, mainly due to the low levels of background. genetic variability of the elite peach materials (Micheletti et al., 2015). In this paper we elaborate a high-density map of an ‘Earlygold’ F2 ‘Earlygold’, an old peach cultivar bred in the US, was crossed with progeny, and examine its variability for a set of characters of agronomic ‘Texas’ (syn. ’Texas prolific’, syn. ’Mission’) almond in the peach root­ interest, to understand their inheritance and to find useful marker-trait stock breeding program of IRTA during the 80 s and produced several associations. We compared these results with those of a QTL analysis on hybrids. One of these, particularly fertile and prolific, was chosen to the BC1 (‘Texas’ × ‘Earlygold’) × ‘Earlygold’ studied by Donoso et al. obtain an F2 progeny to construct the first saturated linkage map of (2016), that allowed us to examine the QTLs of ‘Earlygold’ in two Prunus (Joobeur et al., 1998), which later became the reference for the different genetic backgrounds. These data will also provide information genus (Aranzana et al., 2003; Dirlewanger et al., 2004). The F2 useful to understand the importance of the ‘Earlygold’ allelic variation * Corresponding autor. E-mail address: pere.arus@irta.cat (P. Arús). https://doi.org/10.1016/j.scienta.2021.110726 Received 24 August 2021; Received in revised form 3 October 2021; Accepted 31 October 2021 Available online 17 November 2021 0304-4238/© 2021 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). N. Kalluri et al. Scientia Horticulturae 293 (2022) 110726 in the characters that will be examined in the IL collection currently visually determined by the% of red color at maturity (1 = 0–25%, 2 = under construction. 25–50%, 3 = 50–75% and 4 = 75–100%). Fruit firmness (FF) was evaluated with a hand penetrometer (Wagner, Model 53,200), taking Materials and methods the average value of three fruits with the measurements from both sides for each fruit. Soluble solid content (SSC), expressed in Brix degrees, was Plant materials measured from the juice of three fruits using a digital refractometer (Atago, Tokyo, Japan). Titratable acidity (TA) and pH were determined The ‘Earlygold’ F2 population N = 75 (E × E) was used for the study. using a HI-84,532–02 Titratable Acidity Mini Titrator and a pH meter The trees are located at the IRTA experimental station of Torre Marimon (Hanna instruments, Rhode Island, USA) by diluting 5 ml of fruit juice (Caldes de Montbui, Spain), planted on their own roots in 2015. The with 45 ml of water and titrating with 0.5 M NaOH to a pH of 8.2. TA spacing between the trees was 2.5 m and between the rows 4.0 m. The was calculated in g/l of malic acid. trees were thinned out during the second and third year of fruiting (2018 Leaf: Chlorophyll content (CC) was estimated as the average from ten and 2019). Young leaves were collected from all the trees for DNA leaves per tree using SPAD 502 (Konica Minolta, Osaka, Japan). During extraction (Doyle and Doyle, 1990), to perform genotyping. The E × E the months of July-August, eleven leaves per tree were collected from population initially consisted of 81 plants, later two plants died and four the middle of the tree branches. The leaves were then scanned and their did not produce genotypic data, resulting in 75 plants being used for images stored as TIF files for further analysis. The leaf dimensions were mapping and phenotyping. measured using a Tomato Analyzer 3.0 (http://www.oardc.ohio-state. edu/vanderknaap) software. Leaf parameters analyzed (Fig. 1) were leaf length (LL), petiole length (PL), leaf blade length (LBL), leaf blade Phenotyping width (LBW), leaf shape (LSH) as the ratio of LBL/LBW, leaf perimeter (LP) and leaf surface (LS). All the measurements were in cm, except for The population was evaluated for 24 traits, 18 over three seasons in LSH that is a ratio, and LS that was measured in cm2. Later, the leaves the years 2017 to 2019 and six in only two seasons: chlorophyll content were placed in an incubator for 3 days at 60 ◦ C to determine the leaf dry of the leaves and leaf dry weight (only in 2018 and 2019), leaf color at weight (LW), in grams (g). Leaf color at senescence (LCS) was scored senescence and early and late leaf fall (2019 and 2020), and beginning of visually over two years, once a week, in September and October (Fig. 2) shooting (2020 and 2021). These characters were also analyzed by using a scale of 1 (non-purple, including green yellow and red) and 2 Donoso et al. (2015, 2016), except for leaf color at senescence, fruit (purple leaf). firmness and pH that we analyze here for the first time. The phenotyping The phenotypic data were analyzed statistically using JMP 14.0 methods were essentially identical to those used by these authors. The software (SAS Institute, Cary, NC, USA). Correlations between different characters scored can be classified into four main categories and their traits and years were calculated using the Spearman correlation measurement is described below: coefficient. Flower: The flower shape, showy (large petals) and non-showy (small petals) is determined by a major gene (Sh/sh) in peach, where showy flowers are homozygous (shsh) for the recessive allele (Bailey and Construction of linkage map and QTL analysis French, 1942). Phenology: Flowering time (FT) was scored as the number of Julian For linkage analysis, genotype data were obtained from the 9k In­ days when 50% of the flowers were open. Beginning of shooting (BS) ternational Peach SNP Consortium (IPSC) Illumina Infinium SNP array was the number of Julian days when 5% of the shoots start to appear. (Verde et al., 2012) in 75 plants of the E × E population. The linkage Maturity date (MD) was measured as the number of Julian days with map was constructed using MapMaker/exp 3.0 (Lander et al., 1987) 50% of the fruits mature, as determined by changes in the skin color and using the Kosambi distance function. We initially ordered the markers flesh firmness. Fruit development period (FDP) was scored as the dif­ based on their physical position and established a set of bins (i.e. groups ference in days between the flowering time and maturity date. Begin­ of markers with identical genotype for all the individuals), where each ning of leaf fall (BLF) was scored as the number of Julian days when 10% bin is separated from the adjacent bin by a single or a few recombination of leaves had dropped, and end of leaf fall (ELF) when 90% had dropped. events. Finally, a single SNP from each bin was selected for the dataset Fruit: Fruit weight (FW), in grams (g), was the average weight of six used to construct the linkage map. References for chromosome/linkage mature fruits per individual using a digital balance. Fruit production group numbers and orientation and physical positions were those of (FP) was estimated on a scale of 1 to 4 (1=no fruits, 2= <10 fruits, 3 = Dirlewanger et al. (2004) and Verde et al. (2017). 10–50 fruits and 4= >50 fruits). Intensity of red skin color (ISC) was QTLs for all the traits were analyzed using the interval mapping Fig. 1.. Schematic representation of leaf dimensions (PL petiole length, LBL leaf blade length, LL leaf length, LBW leaf blade width). 2 N. Kalluri et al. Scientia Horticulturae 293 (2022) 110726 Fig. 2.. Leaf color at senescence a, b-yellow, c-green d-red and e,f-purple. method with the MapQTL 6.0 software (Van Ooijen et al. 2002). All the 1640 in E × E and only 1050 in E, resulting in a marker density of 0.27 QTLs with a LOD ≥ 3.0 were considered as significant, as were those cM/marker in E × E compared to 0.47 in the E map. When examining the QTLs with a LOD ≥ 2.5 in one year and a LOD ≥ 3.0 in the rest. The QTLs mapped markers in detail, we observed that their distribution was were considered consistent if they were detected every year. The maps similar in both maps, and that most of the 1050 SNPs of E (97%) were and positions of the QTLs were drawn using the MapChart 2.3 software mapped in E × E, but of the SNPs mapped only in E × E (623) the ma­ (Voorrips, 2002). jority (411; 66%) had a dominant pattern of inheritance, with only two Gene action (GA) was established following the guidelines of genotypes from the three expected (3:1 ratio), instead of the usually Tanksley (1993), based on the values of additivity, a=(A-B)/2, and codominant 1:2:1 SNP inheritance expected in an F2 progeny. dominance d = H-[(A + B)/2], where A and B are the average values of the trait in the homozygous individuals for a given marker in the female Trait phenotypic data and QTL analysis and male parent, respectively, and H that of the heterozygotes. We considered gene action additive (A) if the quotient |d/a| ≤ 0.5, domi­ All of the 24 traits scored were quantitative, except for flower shape nant (D) if 0.5< |d/a|≤1.5, and overdominant (O) if |d/a|>1.5. (Sh), which was scored as qualitative and placed on G8 at the position 23.8 cM with the nearest marker (SNP_IGA_862,006), with physical Results position 13,825,065 bp, cosegregating with the gene. Trait distributions and main features are described in Supplementary Table 2 and Sup­ Linkage maps and map comparisons plementary Figure 2. Those that significantly departed from normal in more than one year were phenology-related (FT, BS, MD, FDP, FP, BLF From the SNPs of the 9k chip, 1640 were polymorphic in the E × E and ELF), leaf color at senescence (LCS) and three fruit characters (ISC, population and distributed in 269 bins. The E × E map covered a total FF and pH). genetic distance of 439.1 cM, detecting the expected eight linkage Correlation analysis was performed between all quantitative traits groups with a higher end of 76.2 cM for linkage group 2 (G2) to a lower and the years in which they were measured (Supplementary Table 3). end of 37.1 cM for G4. The overall physical length covered by this map The correlations between years for a particular trait were high and was 186.7 Mbp, 82.7% of the total physical distance (225.6 Mbp) of the positive for FT, BS, MD, FDP, FW, SSC, CC, and LCS, and intermediate- peach genome v2.0a1 (Verde et al., 2017). There were 24 gaps of to-low for the rest. For correlations between traits, there was a clear >2Mbp, the largest at the distal end of G4 (15.3 Mbp), overall ac­ positive correlation between MD, FDP, FW and SSC, and between BLF counting for 116.4 Mbp, equivalent to 52% of the total physical distance and ELF. Most leaf traits (LP, LS, LBW, LL and LB) of the same year were of the sequenced peach genome (Supplementary Table 1). strongly correlated, but the correlation decreased when comparing data The map of ‘Earlygold’ (the E map) constructed with the (‘Texas’ × from different years. A negative correlation was observed for MD and ‘Earlygold’) × ‘Earlygold’ BC1 population (Donoso et al., 2015) using characters related to anthocyanin coloration of senescent leaves (LCS) the same SNP chip plus 41 SSR markers, is similar in many respects to and intensity of the red skin fruit color (ISC). the one we present here (see comparisons in Supplementary Table 1 and QTLs were detected for 19 traits, between one to four QTLs per trait, Supplementary Figure 1), particularly with respect to the physical and none were identified for four traits (ELF, FP, LBW and LW), giving a coverage (189.7 Mbp), number of bins (214), number of gaps >2 Mbp total of 33 QTLs (Supplementary Table 4 and Supplementary Figure 3). (23; the same as in the E × E map) and physical distance covered by Twelve QTLs of those detected (36%) were consistent in all years studied these gaps (109.0 Mbp). The genetic distance of the E map (521.2 cM) (Table 1, Supplementary Table 4 and Fig. 3). Most of the consistent QTLs was 16% higher than that of the E × E map, and some linkage groups detected explained the variability of phenology (FT, BS, MD and FDP) were clearly longer than in the E × E map (Supplementary Table 1), and fruit (FW, SSC, TA) traits, whereas of the QTLs for the leaf traits, although not significantly longer (the paired t-test of the differences of only CC and LCS were consistent. the genetic distances between the common markers in the extremes of Considering only consistent QTLs (Table 1; Fig. 1), two were found each chromosome was 1.21). The most striking difference between the E for flowering time, one on each of G7 and G8 with R2 = 28.3–42.8% and and E × E maps was the number of SNP markers that could be mapped: R2 = 16.5–20.9% respectively. Also, two QTLs for beginning of shooting, 3 N. Kalluri et al. Scientia Horticulturae 293 (2022) 110726 Table 1 Summary of consistent QTLs detected in the E × E map (‘Earlygold’ F2 population) and E map [‘Earlygold’ map from the (‘Texas’ × ‘Earlygold’) × ‘Earlygold’ population]. Trait category, map type, QTL names, LOD score of the maximum peak, linkage group, map position of the maximum peak, percentage of explained phenotypic variance (R2), additivity (a), dominance (d), d/a, gene action (GA). Trait Map QTL name LOD G Position (cM) R2 a d d/a GA Flowering time E×E qFT7 5.28–8.98 7 30.1–36.9 28.3–42.8 − 2.07 to − 2.75 − 0.11 to − 1.90 0.04 to 0.92 A-D E×E qFT8 2.86–3.78 8 2.8–23.5 16.5–20.9 − 0.12 to 1.91 − 0.85 to 2.50 − 0.45 to − 18.66 A-O End of flowering E1 qFT7 2.50–4.65 7 36.8–40.8 8.9–14.3 3.07 to − 4.24 – – – Beginning of shooting E×E qBS7 5.70–9.27 7 34.9–35.9 30.9–45.2 − 2.84 to − 3.46 − 0.62 to 0.15 − 0.04 to 0.22 A E×E qBS4 3.25–4.42 4 4.3 19.0–24.9 2.06 to 2.30 1.05 to 1.09 0.45 to 0.52 A-D E1 qBS7 4.24–8.13 7 41.4 12.1–23.5 3.72 to 7.03 – – – Maturity date E×E qMD4 24.91–27.14 4 34.3–37.1 80.6–82.0 9.62 to 10.93 − 1.23 to − 5.18 − 0.11 to − 0.51 A-D Fruit development period E×E qFDP4 23.84–30.12 4 37.1 80.1–85.0 9.27 to 11.32 − 1.22 to − 5.26 − 0.10 to − 0.49 A Fruit weight E×E qFW4 4.85–7.73 4 33.6–37.1 28.0–38.6 12.18 to 18.32 − 0.87 to − 6.40 − 0.35 to 0.21 A Soluble solid content E×E qSSC4 8.77–9.51 4 36.4–37.1 44.3–47.4 1.23 to 1.62 − 0.45 to 0.20 − 0.27 to 0.14 A Titratable acidity E×E qTA6 3.04–6.43 6 48.1–55.6 17.4–34.9 − 0.67 to − 0.90 0.09 to 0.24 − 0.13 to − 0.26 A Chlorophyll content E×E qCC3 2.87–3.39 3 46.3–49.0 16.3–19.0 − 0.47 to 0.12 1.15 to 3.28 − 2.45 to 27.33 O Leaf color at senescence E×E qLCS3 5.01–5.19 3 35.3 26.8–28.6 0.30 to 0.33 0.35 to 0.42 1.06 to 1.40 D E×E qLCS4 7.88–8.48 4 36.4–37.1 40.0–41.0 − 0.40 to − 0.41 − 0.006 to 0.01 − 0.02 to 0.02 A 1 Data from the E map are obtained from Donoso et al. (2016). one on each of G7 and G4 with R2 = 30.9–45.2% and R2 = 19.0–24.9%, the analysis of gene action (dominance, additivity, overdominance). F2 respectively. A QTL with a large effect for maturity date (R2 = progenies are also more appropriate for constructing linkage maps, as 80.6–82.0%) and fruit development period (R2 = 80.1–85.0%) was they have more recombination events per meiosis (those of both par­ found at the end of G4, and, at the same position, two other QTLs were ents), so more accurate maps can be constructed with the same number identified for fruit weight and soluble solid content explaining of progeny (Allard 1956). 28.0–38.6% and 44.3–47.4% of the phenotypic variance, respectively. Here we constructed a linkage map with the 9k peach IPGI SNP chip For titratable acidity, a QTL was detected at the end of G6, with R2 in the selfed progeny of ‘Earlygold’, which resulted in a high-density =17.3–34.9%. Three QTLs were identified for leaf-related traits, one for map with coverage of all eight peach chromosomes, with large frag­ chlorophyll content on G3 (R2 =16.3–19.0%) and two for leaf color at ments of DNA without segregating markers, approximately half of the senescence on G3 and G4 explaining 26.8–28.6% and 40.0–41.0% of ‘Earlygold’ genome, suggesting as in previous studies (Eduardo et al., phenotypic variance, respectively. 2013; Martínez-García et al., 2013; Donoso et al., 2015; Serra et al., 2017) that there are large regions of the peach genome identical by Comparison with the QTLs detected in the E map of the T1E population descent owing to the high level of coancestry of the cultivated gene pool. This map was similar in most respects (covered distance, chromosome Only six QTLs were detected on the map of ‘Earlygold’ obtained from length, gaps without markers) to the one obtained for ‘Earlygold’ in the the T1E population by Donoso et al. (2016), all of which we found using interspecific backcross (‘Texas’ × ‘Earlygold’) × ‘Earlygold’ that was the E × E population. Two were consistent QTLs in E, one for end of previously analyzed by Donoso et al. (2015). There was an important flowering time, and the other for beginning of shooting, both located on difference between these two maps, concerning the number of markers G7 at the position of the consistent qFT7 for flowering time in the E × E that could be integrated on the map: 1640 SNPs in E × E, 56% more than map (Table 1). Two additional QTLs, for maturity date and fruit devel­ the 1050 in E. The fact that about two-thirds of the additional markers in opment period, both on G4 in the E map, correspond to the consistent E × E had dominant segregations suggests that a major cause for this qMD4 and qFDP4, respectively, identified here (Table 1). In the E map, difference is that these SNPs could have been discarded in the E map these two latter QTLs were found only in one of the three years studied. because they did not segregate (AA x Aa situations) or that segregations Similarly, beginning of flowering time (BFT) that was mapped to the could have been strongly skewed compared to the two expected (1:1 and same position as qFT7, was detected in only two of the three years 1:2:1), resulting a loss of information compared to what can be obtained studied in E, and a QTL for beginning of shooting in G4 identified in only in an F2 population. While the quality of dominant markers is lower than one of the three years in E probably corresponds to the consistent qBS4 codominant ones for mapping and QTL analysis, the results indicate the in E × E. In all, for a set of 20 common characters, we found six QTLs in E importance of considering all possible segregation types before the and 26 in E × E (see Table 1 and Supplementary Table 4). initial marker filtering steps to maximize the available information for genetic analysis when using large sets of data. The gene for flower type, Sh, was found as expected in G8 and at a Discussion similar map position as in Ogundiwin et al. (2009), who mapped this gene for the first time, and as in Donoso et al. (2016). Its physical po­ Due to the self-compatibility system of peach, it is possible to obtain sition on chromosome 8 is also compatible with the results of Micheletti F2 progenies as well as F1 or backcross progenies for genetic analysis. F1 et al. (2015) and Cao et al. (2016), both using genome-wide association segregating progenies are frequently used for QTL analysis as they are analysis. the usual progeny type used in peach breeding. Since the parents used A major QTL (qMD4) explaining >80% of the phenotypic variance are partly heterozygous lines or cultivars, they segregate in the F1 for maturity date was found on G4. A QTL in this position has been progeny. These progenies, also termed pseudo-backcrosses (Grattapa­ detected by other authors in a very broad transect of peach materials as glia and Sederoff 1994), are generally analyzed as two different back­ well as in other Prunus species (Dirlewanger et al., 2012; Hernández cross one (BC1) populations, each corresponding to one of the parents of Mora et al. 2017; Serra et al., 2017; Aranzana et al., 2019; Rawandoozi the cross, and linked by markers that are heterozygous in both parents et al., 2021; Chen et al., 2021). In certain cases, this trait has been in­ and then segregate 1:2:1 (or 1:1:1:1 when using multi-allelic markers). tegrated in the maps as a major gene MD/md (Eduardo et al., 2011; F2 populations are also used, especially to better understand the inher­ Pirona et al., 2013). A strong candidate gene for this character is an NAC itance of specific characters as they recover the three possible genotypes, transcription factor (Prupe.4G186800.1), orthologous to the Nor gene in unlike BC1 populations that recover only two, making them suitable for 4 N. Kalluri et al. Scientia Horticulturae 293 (2022) 110726 Fig. 3.. Map of the ‘Earlygold’ F2 progeny with the positions of a major gene (Sh) and 12 consistent QTLs over the years. Colors of the bars of QTLs are as follows: pink flower, red fruit and green leaf traits. 5 N. Kalluri et al. Scientia Horticulturae 293 (2022) 110726 tomato (Pirona et al., 2013). The same region includes a gene (Sr/sr) While the QTL on G3 has been generally detected as the main deter­ responsible for another character, slow ripening, that determines the minant in the color-related traits studied in Prunus, it is often accom­ presence of individuals producing fruit that do not ripen, remaining panied by a QTL on G4, corresponding to the position of qLCS4 found immature on the tree for a long time (Eduardo et al., 2015; Núñez-Lillo here, such as for skin color in peach, almond × peach and Japanese plum et al., 2015). A 26.6 kb deletion containing the Prupe.4G186800.1 gene crosses (Frett et al., 2014; Donoso et al., 2016; Salazar et al., 2017; was associated with the sr allele (responsible for the slow ripening Hernández Mora et al. 2017) and flesh color in cherry (Calle et al., phenotype) suggesting that MD and Sr may be the same gene (Eduardo 2021). The inheritance of non-anthocyanic vs. anthocyanic senescing et al., 2015; Meneses et al., 2016). Another major gene mapped in this leaves in E × E appears to have an oligogenic inheritance, based on the region is Alf/alf that determines the formation of the thick mesocarp combination of genotypes of these two QTLs and their interaction. characteristic of the peach fruit in almond × peach progenies (Donoso qLCS3 had a dominant gene action and qLCS4 was additive (see Table 1) et al., 2016). Additionally, this region of chromosome 4 contains several with alleles B and A, respectively, being responsible for absence of QTLs involved in the inheritance of other fruit traits, including FDP, anthocyanin color. Individuals having one of these alleles or both in which can be considered as an alternative measurement of MD homozygosis, selected using the SNPs closest to the LOD peaks of the (Hernández Mora et al. 2017), and many other characters including SSC, QTLs (SNP_IGA_344,086 for qLCS3 and SNP_IGA_405,773 for qLCS4), fruit weight, acidity, pH and red skin color (Eduardo e al. 2011, were usually (90.2% of the cases) non-colored or anthocyanic, as pre­ Hernández Mora et al. 2017; Rawandoozi et al., 2021). It is unknown dicted by the markers. whether the concurrence of genes involved in the inheritance of so many One remarkable finding of this research is that using an F1 popula­ characters in this region is due to the action of a highly polymorphic tion, i.e. with the ‘Earlygold’ QTLs studied using a backcross type single gene with a broad set of pleiotropic effects, or to a cluster of genes segregation, we only found a subset (6) of the 26 QTLs that were that would constitute a hotspot for a diverse set of characters in peach detected using an F2 progeny. There are two reasons for this important and other Prunus. In this work, we identified consistent QTLs for four difference. First, trait segregation between peach and almond, corre­ characters in this region, including FDP, SSC, FW and LCS: in all cases, sponding to QTLs heterozygous in the ‘Texas’ × ‘Earlygold’ hybrid late maturity was associated with an increase of these traits except for parent in the F1 progeny, produced often by alleles with greater relative leaf color at senescence where late maturity and presence of anthocy­ effects than those segregating within ‘Earlygold’, may have interfered anin color were negatively correlated. The QTLs of all traits at this re­ with the identification of QTLs at the same or different genomic loca­ gion were essentially additive (Table 1) and with a generally large effect tions, resulting in a loss of efficiency in the detection of ‘Earlygold’ QTLs. (R2=28.0–85.0%). Overall, this region appears to have a major impact And second, heterozygous QTLs in both ‘Earlygold’ and the ‘Texas’ × on the phenotype and is usually the determinant of critical aspects of ‘Earlygold’ hybrid parent would segregate 3:1 or 1:2:1 in the progeny, agronomic characters that are affected by phenology and fruit-related but they were analyzed with markers that segregated 1:1, resulting in a traits. It deserves detailed exploration at the genetic level to under­ reduction of power to identify QTLs; heterozygous individuals for the stand its nature and how different haplotypes may shape expression of QTL cannot be used for genetic analysis in this case, halving the effective the traits involved in this highly polymorphic region. population size. While F1 progenies between partly heterozygous par­ Two consistent QTLs found for flowering time (qFT7) and beginning ents are often used for QTL analysis in clonally propagated species, a of shooting (qBS7) mapped at the same location, suggesting that they more efficient QTL analysis can be done with other population types could correspond to the same locus. These two characters differed in two (particularly F2 progenies). In addition, F2 populations allow for the QTLs, qBS4, located at a different position to the MD-related QTLs, and analysis of QTL action (dominance, additivity and overdominance), qFT8, indicating that these characters may be encoded by a partly which is not possible with backcross populations. One possible way to overlapping set of genes. Their positions also coincide with those found rescue as much information as possible from F1 segregating progenies is in peach and other Prunus species (Fan et al., 2010; Silva et al., 2005; analyzing QTLs in two steps, the first one as a backcross, followed by a Dirlewanger et al., 2012; Hernández Mora et al. 2017) where they have second analysis using only 1:2:1 segregating markers. The latter analysis been studied along with a cohort of other QTLs that depend on the would detect F2 segregating QTLs with more precision, as the informa­ specific population studied. tion from both parents will be used, resulting in an increase of their LOD We identified a consistent QTL for fruit juice titratable acidity (TA) values in addition to the possible identification of new QTLs not on G6 (qTA6), at a different position of the major gene determining significantly detectable when using a BC1 progeny for the analysis. subacid vs. acid fruit (D/d) that maps on G5 (Dirlewanger et al., 2006) One of the reasons for the analysis of the segregation of ‘Earlygold’ is and is homozygous for the allele (d) that confers fruit acidity in ‘Ear­ that this cultivar has been used as the recurrent parent for the intro­ lygold’. This QTL was already detected with strong evidence in the gression line collection peach/almond currently under construction. An multiple progeny analysis performed by Hernández Mora et al. (2017), inbred line is typically used for that in the IL collections available, but in confirming its value as a factor that may be used to modulate fruit the absence of peach inbred lines of the major commercial gene pool acidity in the absence of the subacid allele. with sufficient vigor, we opted for ‘Earlygold’ which, as we have shown The leaf color at senescence was analyzed here for the first time in here and in previous studies (Donoso et al., 2015), has large homozy­ peach and we found two loci, both with strong effects (R2=28.6–40.5%), gous regions adding up to approximately half of its genome. Knowing one on G4 (qLCS4) and the other on G3 (qLCS3). A major gene for red vs. the segregating QTLs of this genotype is important as the IL collection green leaf color (Gr/gr) has been described in peach and mapped to G6 will be segregating for them, and their variability may interfere with the (Lambert and Pascal, 2011), which discards its involvement in the analysis of the characters segregating between peach and almond, which variability observed in the ‘Earlygold’ progenies. The position of qLCS3 are the relevant ones in this case. Our results indicate that there is corresponds to that of the gene that determines yellow vs. anthocyanin essentially one region of concern, that of chromosome 4 that contains anther color (Ag/ag; Arús et al., 1994), the “highlighter” gene that the QTLs related with maturity date and affects many other fruit traits. controls peach fruit skin red blush (H/h; Bretó et al., 2017), and to Having this region with a genetic composition that results in a similar various QTLs determining anthocyanin color in fruit skin and flesh of expected phenotype for all plants of the IL collection is a clear conclu­ several stone fruit crops (Sooriyapathirana et al., 2010; Socquet-Juglard sion from the data presented here. et al., 2013; Donoso et al., 2016; Calle et al., 2021; Fiol et al., 2021). The phenotypic variability of these genes in Prunus has been attributed to the CRediT authorship contribution statement variation of one or several tandemly-duplicated transcription factor genes (PpMYB10.1, PpMYB10.2 and PpMYB10.3) involved in the Naveen Kalluri: Methodology, Investigation, Formal analysis, anthocyanin metabolic pathway (Tuan et al., 2015, Fiol et al.2021). Writing – original draft. Iban Eduardo: Funding acquisition, Resources, 6 N. Kalluri et al. Scientia Horticulturae 293 (2022) 110726 Writing – review & editing. Pere Arús: Conceptualization, Funding compounds from the essential oil of peach fruit: QTL analysis and identification of candidate genes using dense SNP maps. Tree Genet. Genomes 9, 189–204. acquisition, Writing – review & editing, Supervision. Eduardo, I., Picañol, R., Rojas, E., Batlle, I., Howad, W., Aranzana, M.J., Arús, P., 2015. Mapping of a major gene for the slow ripening character in peach: co-location with the maturity date gene and development of a candidate gene-based diagnostic Declaration of Competing Interest marker for its selection. Euphytica 205, 627–636. Fan, S., Bielenberg, D.G., Zhebentyayeva, T.N., Reighard, G.L., Okie, W.R., Holland, D., The authors declare that they have no known competing financial Abbott, A.G., 2010. Mapping quantitative trait loci associated with chilling interests or personal relationships that could have appeared to influence requirement, heat requirement and bloom date in peach (Prunus persica). New Phytologist 185, 917–930. https://doi.org/10.1111/j.1469-8137.2009.03119.x. the work reported in this paper. Fiol, A., García-Gómez, B.E., Jurado-Ruiz, F., Alexiou, K., Aranzana, M.J., 2021. Characterization of Japanese plum (Prunus salicina) PsMYB10 gene alleles reveals structural variation and polymorphisms correlating with fruit skin color. Front. Plant Acknowledgments Sci. 12, 655267 https://doi.org/10.3389/fpls.2021.655267. Frett, T.J., Reighard, G.L., Okie, W.R., Gasic, K., 2014. Mapping quantitative trait loci This research has been partly funded by grant PID2019–110599RR- associated with blush in peach [Prunus persica (L.) Batsch]. Tree Genet. Genomes 10, 367–381. https://doi.org/10.1007/s11295-013-0692-y. I00 from the Spanish Ministry of Science and Innovation, by the Euro­ Grattapaglia, D., Sederoff, R., 1994. Genetic linkage maps of Eucalyptus grandis and pean Union Program PRIMA (project FREECLIMB no. Eucalyptus urophylla using a pseudo-testcross: mapping strategy and RAPD markers. PCI2019–103670), Severo Ochoa Program for Centres of Excellence in Genetics 137, 1121–1137. Hernández Mora, J.R., Micheletti, D., Bink, M., van de Weg, E., Cantín, C., Nazzicari, N., R&D 201–2019 SEV-2015–0533 and CERCA Program-Generalitat de Caprera, A., Detori, M.T., Micali, S., Banchi, E., Campoy, J.A., Dirlewanger, E., Catalunya. Lambert, P., Pascal, T., Troggio, M., Bassi, D., Rossini, L., Verde, I., Quilot-Turion, B., Laurens, F., Arús, P., Aranzana, M.J., 2017. Integrated QTL detection for key breeding traits in multiple peach progenies. BMC Genomics 18, 404. Supplementary materials Joobeur, T., Viruel, M.A., de Vicente, M.C., Jáuregui, B., Ballester, J., Dettori, M.T., Verde, I., Truco, M.J., Messeguer, R., Batlle, I., Quarta, R., Dirlewanger, E., Arús, P., Supplementary material associated with this article can be found, in 1998. Construction of a saturated linkage map in Prunus using an almond x peach F2 progeny. Theor. Appl. Genet. 97, 1034–1041. the online version, at doi:10.1016/j.scienta.2021.110726. Lambert, P., Pascal, T., 2011. Mapping Rm2 gene conferring resistance to the green peach aphid (Myzus persicae Sulzer) in the peach cultivar “Rubira®. Tree Genet. References Genomes 7, 1057–1068. https://doi.org/10.1007/s11295-011-0394-2. Lander, E.S., Green, P., Abrahamson, J., Barlow, A., Daly, M.J., SE, Lincoln, Newburg, L., 1987. MAPMAKER: an interactive computer package for constructing primary Allard, R.W., 1956. Formulas and tables to facilitate the calculation of recombination genetic linkage maps of experimental and natural populations. Genomics 1, values in heredity. Hilgardia 24 (10), 235–278. 174–181. https://doi.org/10.1016/0888-7543(87)90010-3. Aranzana, M.J., Pineda, A., Cosson, P., Dirlewanger, E., Ascasibar, J., Cipriani, G., Martínez-García, P.J., Parfitt, D.E., Ogundiwin, E.A., Fass, J., Chan, H.M., Ahmad, R., Ryder, C.D., Testolin, R., Abbott, A.G., King, G.J., Iezzoni, A.F., Arús, P., 2003. A set Lurie, S., Dandekar, A., Gradziel, T.M., Crisosto, C.H., 2013. High density SNP of simple-sequence repeat (SSR) markers covering the Prunus genome. Theor. Appl. mapping and QTL analysis for fruit quality characteristics in peach (Prunus persica Genet. 106, 819–825. https://doi.org/10.1007/s00122-002-1094-y. L.). Tree Genet. Genomes 9, 19–36. Aranzana, M.J., Decroocq, V., Dirlewanger, E., Eduardo, I., Gao, Z.S., Gasic, K., Meneses, C., Ulloa-Zepeda, L., Cifuentes-Esquivel, A., Infante, R., Cantin, C., Batlle, I., Iezzoni, A., Jung, S., Peace, C., Prieto, H., Tao, R., Verde, I., Abbott, A.G., Arús, P., Arús, P., Eduardo, I., 2016. A codominant diagnostic marker for the slow ripening 2019. Prunus genetics and applications after de novo genome sequencing: trait in peach. Mol. Breed. 36, 77. achievements and prospects Horticulture Research 6, 58. Micheletti, D., Dettori, M.T., Micali, S., Aramini, V., Pacheco, I., da Silva, C., Foschi, S., Arús, P., Messeguer, R., Viruel, M., Tobutt, K., Dirlewanger, E., Santi, F., Quarta, R., Banchi, E., Barreneche, T., Quilot-Turion, B., Lambert, P., Pascal, T., Iglesias, I., Ritter, E., 1994. The European Prunus mapping project. Progress in the almond Carbó, J., Wang, L.R., Ma, R.J., Li, X.W., Gao, Z.S., Nazzicari, N., Troggio, M., linkage map. Euphytica 77, 97–100. Bassi, B., Rossini, L., Verde, I., Laurens, F., Arús, P., Aranzana, M.J., 2015. Whole- Bailey, J.S., French, A.P., 1942. The inheritance of blossom type and blossom size in genome analysis of diversity and SNP-major gene association in peach germplasm. peach. Proc. Am. Soc. Hort. Sci. 40, 248–250. PLoS ONE 10 (9), e0136803. Bretó, M., Cantín, C., Iglesias, I., Arús, P., Eduardo, I., 2017. Mapping a major gene for Núñez-Lillo, G., Cifuentes-Esquivel, A., Troggio, M., Micheletti, D., Infante, R., Campos- red skin color suppression (highlighter) in peach. Euphytica 213 (1), 14. Vargas, R., Orellana, A., Blanco-Herrera, F., Meneses, C., 2015. Identification of Calle, A., Serradilla, M.J., Wünsch, A., 2021. QTL mapping of phenolic compounds and candidate genes associated with mealiness and maturity date in peach [Prunus fruit colour in sweet cherry using a 6+9K SNP array genetic map. Sci. Hortic. 280, persica (L.) Batsch] using QTL analysis and deep sequencing. Tree Genet. Genomes 109900 https://doi.org/10.1016/j.scienta.2021.109900. 11, 13. DOI: 86 10.1007/s11295-015-0911-9. Cao, K., Zhou, Z., Wang, Q., Guo, J., Zhao, P., Zhu, G., Fang, W., Chen, C., Wang, X., Ogundiwin, E.A., Peace, C.P., Gradziel, T.M., Parfitt, D.E., Bliss, F.A., Crisosto, C.H., Wang, X., Tian, Z., Wang, L., 2016. Genome-wide association study of 12 agronomic 2009. A fruit quality gene map of Prunus. BMC Genomics 10, 587. traits in peach. Nat. Commun. 7, 1–10. https://doi.org/10.1038/ncomms13246. Pirona, R., Eduardo, I., Pacheco, I., Linge, C.D., Miculan, M., Verde, I., Tartarini, S., Chen, C., Guo, J., Cao, K., Zhu, G., Fang, W., Wang, X., Li, Y., Wu, J., Xu, Q., Wang, L., Dondini, L., Pea, G., Bassi, D., Rossini, L., 2013. Fine mapping and identification of a 2021. Identification of candidate genes associated with slow-melting flesh trait in candidate gene for a major locus controlling maturity date in peach. BMC Plant Biol. peach using bulked segregant analysis and RNA-seq. Scientia. Hortic. 286, 110208. 13, 166. https://doi.org/10.1186/1471-2229-13-166. Dirlewanger, E., Graziano, E., Joobeur, T., Garriga-Calderé, F., Cosson, P., Howad, W., Rawandoozi Z.J. , Hartmann T.P., Carpenedo S., Gasic K. , da Silva Linge C., Cai L.C. , Arús, P., 2004. Comparative mapping and marker assisted selection in Rosaceae fruit Van de Weg E., Byrne D.H. (2021) Mapping and characterization QTLs for crops. Proc. Natl Acad. Sci. U.S.A. 101, 9891–9896. https://doi.org/10.1073/ phenological traits in seven pedigree-connected peach famílies. 22: 187. doi: pnas.0307937101. 10.1186/s12864-021-07483-8. Dirlewanger, E., Cosson, P., Boudehri, K., Renaud, C., Capdeville, G., Tauzin, Y., Salazar, J.A., Pacheco, I., Shinya, P., Zapata, P., Silva, C., Aradhya, M., Velasco, D., Laigret, F., Moing, A., 2006. Development of a second-generation genetic linkage Ruiz, D., Martinez-Gomez, P., Infante, R., 2017. Genotyping by sequencing for SNP- map for peach [Prunus persica (L.) Batsch] and characterization of morphological based linkage analysis and identification of QTLs linked to fruit quality traits in traits affecting flower and fruit. Tree Genet. Genomes 3, 1–13. Japanese plum (Prunus salicina Lindl.). Front. Plant Sci. 8, 476. DOI 10.3389/ Dirlewanger, E., Quero-García, J., Le Dantec, L., Lambert, P., Ruiz, D., Dondini, L., fpls.2017.00476. Illa, E., Quilot-Turion, B., Audergon, J.M., Tartarini, S., Letourmy, P., Arús, P., 2012. Serra, O., Donoso, J.M., Picañol, R., Batlle, I., Howad, W., Eduardo, I., Arús, P., 2016. Comparison of the genetic determinism of two key phenological traits, flowering and Marker-assisted introgression (MAI) of almond genes into the peach background: a maturity dates, in three Prunus species: peach, apricot and sweet cherry. Heredity fast method to mine and integrate novel variation from exotic sources in long Heredity (Edinb) 109, 280–292. intergeneration species. Tree Genet. Genomes 12, 96. Donoso, J.M., Picañol, R., Eduardo, I., BatlIe, I., Howad, W., Aranzana, M.J., Arús, P., Serra, O., Giné-Bordonaba, J., Eduardo, I., Bonany, J., Echeverría, G., Larrigaudière, G., 2015. High-density mapping suggests a cytoplasmic male sterility system with two Arús, P., 2017. Genetic analysis of the slow melting flesh character in peach. Tree restorer factors in almond x peach progenies. Hortic. Res. 2, 15016. Genet. Genomes 13, 77. Donoso, J.M., Picañol, R., Serra, O., Howad, W., Alegre, S., Arús, P., Eduardo, I., 2016. Shulaev, V., Korban, S.S., Sosinski, B., Abbott, A.G., Aldwinckle, H.S., Folta, K.M., Exploring almond genetic variability useful for peach improvement: mapping major Iezzoni, A., Main, D., Arús, P., Dandekar, A.M., Lewers, K., Brown, S.K., Davis, T.M., genes and QTLs in two inter-specific almond × peach populations. Mol. Breed. 36, Gardiner, S.E., Potter, D., Veilleux, R.E., 2008. Multiple models for Rosaceae 16. genomics. Plant Phys. 147, 985–1003. Doyle, J.J., Doyle, J.L., 1990. A rapid total DNA preparation procedure for fresh plant Silva, C., Garcia-Mas, J., Sánchez, A.M., Arús, P., Oliveira, M.M., 2005. Looking into tissue. Focus Focus (Madison) 12 (1), 13–15. flowering time in almond (Prunus dulcis (Mill) D. A. Webb): the candidate gene Eduardo, I., Pacheco, I., Chietera, G., Bassi, D., Pozzi, C., Vecchietti, A., Rossini, L., 2011. approach. Theor. Appl. Genet. 110 (5), 959–968. QTL analysis of fruit quality traits in two peach intraspecific populations and Socquet-Juglard, D., Christen, D., Devènes, G., Gessler, C., Duffy, B., Patocchi, A., 2013. importance of maturity date pleiotropic effect. Tree Genet. Genomes 7, 323–335. Mapping architectural, phenological, and fruit quality QTLs in apricot. Plant Mol. Eduardo, I., Chietera, G., Pirona, R., Pacheco, I., Troggio, M., Banchi, E., Bassi, D., Biol. Rep. 31 (2), 387–397. https://doi.org/10.1007/s11105-012-0511-x. Rossini, L., Vecchietti, A., Pozzi, C., 2013. Genetic dissection of aroma volatile 7 N. Kalluri et al. Scientia Horticulturae 293 (2022) 110726 Sooriyapathirana, S.S., Khan, A., Sebolt, A.M., Wang, D., Bushakra, J.M., Lin-Wang, K., Verde, I., Abbott, A.G., Scalabrin, S., Jung, S., Shu, S., Marroni, F., Zhebentyayeva, T., Allan, A.C., Gardiner, S.E., Chagné, D., Iezzoni, A.F., 2010. QTL analysis and Dettori, M.T., Grimwood, J., Cattonaro, F., Zuccolo, A., Rossini, L., Jenkins, J., candidate gene mapping for skin and flesh color in sweet cherry fruit (Prunus avium Vendramin, E., Meisel, L.A., Decroocq, V., Sosinski, B., Prochnik, S., Mitros, T., L.). Tree Genet. Genomes 6 (6), 821–832. https://doi.org/10.1007/s11295-010- Policriti, A., Cipriani, G., Dondini, L., Ficklin, S., MGoodstein, D., Xuan, P., Del 0294-x. Fabbro, C., Aramini, V., Copetti, D., Gonzalez, S., Horner, D.S., Falchi, R., Lucas, S., Tanksley, S.D., 1993. Mapping polygenes. Annu. Rev. Genet. 27, 205–233. Mica, E., Maldonado, J., Lazzari, B., Bielenberg, D., Pirona, R., Miculan, M., Tuan, P.A., Bai, S., Yaegaki, H., Tamura, T., Hihara, S., Moriguchi, T., Oda, K., 2015. The Barakat, A., Testolin, R., Stella, A., Tartarini, S., Tonutti, P., Arús, P., Orellana, A., crucial role of PpMYB10.1 in anthocyanin accumulation in peach and relationships Wells, C., Main, D., Vizzotto, G., Silva, H., Salamini, F., Schmutz, J., Morgante, M., between its allelic type and skin color phenotype. BMC Plant Biol. 15, 280. Rokhsar, D.S., 2013. The high-quality draft genome of peach (Prunus persica) Van Ooijen, J.W., Boer, M.P., Jansen, R.C., Maliepaard, C., 2002. MapQTL® 4.0. identifies unique patterns of genetic diversity, domestication and genome evolution. Software for the calculation of QTL positions on genetic maps. Plant Res. Int. Nat. Genet. 45 (5), 487–494. Wageningen, Netherlands. Verde, I., Jenkins, J., Dondini, L., Micali, S., Pagliarani, G., Vendramin, E., Paris, R., Verde, I., Bassil, N., Scalabrin, S., Gilmore, B., Lawley, C.T., Gasic, K., Micheletti, D., Aramini, V., Gazza, L., Rossini, L., Bassi, D., Troggio, M., Shu, S., Grimwood, J., Rosyara, U.R., Cattonaro, F., Vendramin, E., Main, D., Aramini, V., Blas, A.L., Tartarini, S., Dettori, M.T., Schmutz, J., 2017. The Peach v2.0 release: high- Mockler, T.C., Bryant, D.W., Wilhelm, L., Troggio, M., Sosinski, B., Aranzana, M.J., resolution linkage mapping and deep resequencing improve chromosome-scale Arús, P., Iezzoni, A., Morgante, M., Peace, C., 2012. Development and evaluation of a assembly and contiguity. BMC Genomics 18, 225. https://doi.org/10.1186/s12864- 9k SNP array for peach by internationally coordinated snp detection and validation 017-3606-9. in breeding germplasm. PLoS ONE 7 (4), e35668. https://doi.org/10.1371/journal. Voorrips, R.E., 2002. Mapchart: software for the graphical presentation of linkage maps pone.0035668. and QTLs. J. Heredity 93 (1), 77–78. https://doi.org/10.1093/jhered/93.1.77. 8